提出自适应图卷积机制，让视觉图网络更高效准确
AdaptViG: Adaptive Vision GNN with Exponential Decay Gating
- 用动态门控机制按特征相似性筛选长程连接
- 参数少80%、计算量少84%，准确率仍超现有模型0.3%
- 适合追求高效高精度视觉模型的研究者与工程师
视觉图神经网络（ViGs）为视觉架构发展提供了新方向。尽管强大，ViGs常因图构建阶段带来巨大计算开销，影响效率。为此，我们提出AdaptViG，一种高效且强大的混合视觉图网络，引入新型图构建机制——自适应图卷积。该机制基于高效静态轴向骨架和动态内容感知的指数衰减门控策略，根据特征相似性选择性加权长程连接。此外，AdaptViG采用混合策略：早期使用高效门控机制，最终阶段采用全图注意力块以实现最大特征聚合。方法在准确率与效率之间达到新最优平衡。例如，AdaptViG-M达到82.6%的top-1准确率，优于ViG-B 0.3%，同时参数减少80%，GMACs降低84%。在下游任务中，AdaptViG-M获得45.8 mIoU、44.8 APbox、41.1 APmask，分别超过更大模型EfficientFormer-L7 0.7、2.2、2.1，且参数减少78%。
原文摘要 · Abstract (English)
Vision Graph Neural Networks (ViGs) offer a new direction for advancements in vision architectures. While powerful, ViGs often face substantial computational challenges stemming from their graph construction phase, which can hinder their efficiency. To address this issue we propose AdaptViG, an efficient and powerful hybrid Vision GNN that introduces a novel graph construction mechanism called Adaptive Graph Convolution. This mechanism builds upon a highly efficient static axial scaffold and a dynamic, content-aware gating strategy called Exponential Decay Gating. This gating mechanism selectively weighs long-range connections based on feature similarity. Furthermore, AdaptViG employs a hybrid strategy, utilizing our efficient gating mechanism in the early stages and a full Global Attention block in the final stage for maximum feature aggregation. Our method achieves a new state-of-the-art trade-off between accuracy and efficiency among Vision GNNs. For instance, our AdaptViG-M achieves 82.6% top-1 accuracy, outperforming ViG-B by 0.3% while using 80% fewer parameters and 84% fewer GMACs. On downstream tasks, AdaptViG-M obtains 45.8 mIoU, 44.8 APbox, and 41.1 APmask, surpassing the much larger EfficientFormer-L7 by 0.7 mIoU, 2.2 APbox, and 2.1 APmask, respectively, with 78% fewer parameters.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。